Location: IHERA-Drexel December 2013
Introduction. A cholera outbreak was reported in Haiti in October 2010, and both weather conditions and UN Peace-Keeping Forces were implicated. An outbreak of cholera was also reported in a jute mill in Kolkatta, West Bengal, India that same year. This case study focuses on the Haiti outbreak in an attempt to guide future assessment and control efforts, with comparisons to the Kolkatta outbreak made throughout the paper.
Hazard ID. The V. cholerae strain identified in both outbreaks was the serotype O1, biotype El Tor, Ogawa strain.
Exposure Assessment. Cholera is found naturally in marine and freshwater environments and has a fecal-oral transmission route. Contaminated water is usually the source of infection. Infected individuals can shed up to 10^7 to 10^9 organisms/mL.
Dose-Response. Four different dose-response models were discussed and evaluated as to their validity in the two cholera outbreaks.
Risk Characterization. One of the greatest factors determining risk characterization in Haiti’s outbreak remains with the various modes of transmission. Describing risk in the Haitian outbreak is limited by the dose-response model applicability and the grand extent of the outbreak, which provided multiple complicating factors that were, by contrast, relatively limited in the jute mill in India.
Risk Management. Diagnosis of cholera presents challenges but there are simple and effective methods of treatment. Risk factors, personnel, and prevention methods are described, followed by a discussion of the limitation of these methods as well as the global health implications of the Haiti outbreak.
In October 2010, 10 months after a severely damaging earthquake, an outbreak of cholera was reported in Haiti. The outbreak was first reported in the Center Department, a rural area of Haiti, but soon spread to the city of St. Marc (Artibonite Department). Eventually the country’s remaining departments were also affected. The spread to all 10 departments of Haiti occurred over approximately 1 month. Cholera had not been reported in Haiti, up until this outbreak, in 100 years (Frerichs, Keim, Barrais, & Piarroux, 2012; Gaudart, Rebaudet, Barrais, Boncy, Faucher, et al., 2013).
There were two suggested hypotheses concerning the source of Haiti’s cholera. The first hypothesis suggested that the UN Peace-Keeping Force, sent directly from Nepal to Haiti, was responsible, as Nepal faced an outbreak of cholera between August 2010 and October 2010, the same time that the Peace-Keeping Force was present in Nepal. The second hypothesis suggested that the El Nino phenomenon encouraged naturally-occurring cholera to proliferate in the environment (Frerichs, et al., 2012).
Nepalese soldiers arrived in Meille (near the town of Mirebelais) on October 9, 12, and 16, 2010, and camped above a stream that flowed into the Artibonite River. An investigation of the camp showed an inappropriate sanitation infrastructure, such as a pipe that discharged sewage into the Artibonite River. The waters of the Artibonite River are used for cooking and drinking, and due to repair work being done on the local water supply network damaged in the earthquake, residents of Mirebalais were drinking the water from the river (Piarroux, Barrais, Faucher, Haus, Piarroux, et al., 2011; Gaudart, et al., 2013).
During the outbreak probable cholera cases were defined as anyone reporting profuse acute watery diarrhea. The first case was reported on October 14, 2010 from a family living in the town of Meille, and during that week many more cases were reported in Mirebalais. Environmental and water samples, taken around the 19th of October, demonstrated negative results (Barzilay, Schaad, Magloir, Mung, Boncy, Dahourou, et al., 2013).
After October 19, 2010, many more cases of cholera were reported in the Artibonite Department, located downstream of the river. On October 22, 2010, 14 other areas bordering the Artibonite Department, as well as the city of Port-au-Prince, reported cases of cholera; these cases appeared not long after residents of affected areas attempted to flee to these unaffected areas (Gaudart, et al., 2013).
Many hospital admissions and deaths occurred in the Haiti cholera outbreak, a disease which remains a problem in the country. There were 7526 deaths out of 604,624 cases reported between October 2010 and October 2012 (Barzilay, et al., 2013).
In this same year, between March 6 and March 16, a jute mill in Kolkata, West Bengal, India reported a cholera outbreak. 197 cases presented, all of which were in males between the ages of 15 and 64, but no deaths occurred. The outbreak was determined to have come from a drinking-water reservoir that was contaminated. Unlike Haiti, cholera is endemic to India. Although of the same cholera strain as that of Haiti's epidemic, this outbreak did not have near so long or devastating an impact in Kolkata, as the Indian outbreak was resolved within 10 days (Mridha, Biswas, Ramakrishnan, & Murhekar, 2011).
This case study attempts to provide a framework for a cholera outbreak of a magnitude like that seen in Haiti, with efforts to provide guidance, through example and discussion, in regard to dose-response models, exposure assessment, risk characterization, and risk management. Comparisons between the jute mill outbreak and Haiti outbreak will be made throughout the paper.
Cholera is a disease caused by a motile, gram-negative, curved rod-shaped bacterium that is approximately 0.5 to 1.5 by 3.0 micrometers in size (bacterium Vibrio cholera). There are nearly 200 serologic types, but it is the serologic type O1 that is of concern for epidemics. The type O1 “has been responsible for yearly epidemics in the Indian subcontinent for centuries and, for at least the last two centuries, periodic global pandemics” (Engelberg, DiRita, & Dermody, 2013; Murray, Rosenthal, & Pfaller, 2009). There are two biotypes of the O1 serotype: classical and El Tor. The O1 El Tor biotype was the cause of a pandemic starting in Indonesia in the 1960s, spread to Europe and Africa in the 1970s, then Latin America in 1991; this resulted in cholera becoming endemic in Africa and Latin America. There is an endemic area along the Gulf Coast of the US, but this area is small and only a few cases result each year due to contaminated shellfish (Murray, et al., 2009; Engelberg, et al., 2013). Within both biotypes are three strains: Ogawa, Inaba, and Hikojima. The strain identified in the Haiti cholera epidemic of 2010 was the serotype O1, biotype El Tor, Ogawa strain form of V. cholera; this same strain was responsible for the outbreak in the jute mill in India that same year (Murray, et al., 2009; Parsi, 2001; Barzilay, Schaad, Magloire, Mung, Boncy, et al., 2013; Piarroux, Barrais, Faucher, Haus, Piarroux, et al., 2011; Mridha, Biswas, Ramakrishnan, & Murhekar, 2011).
Virulence, Symptoms, Mortality
The main virulence factor of V. cholera is the cholera toxin, which is released by the bacteria, although there are other virulence factors. The A subunit of cholera toxin is internalized into the intestinal cells causing the secretion of water and electrolytes—this leads to the voluminous, watery diarrhea that is characteristic of the disease: with the increased loss of fluid eventually comes a change in stools to a colorless, odorless fluid without protein and flecked with mucus (“rice-water stools”). “Severely infected patients can lose as much as 1 liter of fluid per hour during the height of the disease”; children may experience loss of fluid up to 10 mL/kg/hr (Murray, et al., 2009, p. 218; Parsi, 2001; Sack et al., 2004). Fortunately, approximately 75% of individuals infected with V. cholera O1 are asymptomatic, while approximately 20% have a self-limited diarrhea. Cholera can cause rapidly-fatal diarrhea in approximately 5 to 7% of infected (WHO, 2012; CDC, 2010; Murray, et al., 2009). The incubation time for O1 El Tor Ogawa V. cholera ranges from 0.2 to 7.4 days, with the median time being approximately 1.3 days (Azman, Rudolph, Cummings, & Lessler, 2013). Patients can become severely dehydrated eventually experiencing tachycardia, hypotension, tachypnea, poor skin turgor, dry mucous membranes, and change in mental status; if treatment does not occur, this may develop to hypovolemic shock, metabolic acidosis, renal failure, arrhythmias, seizure, coma, and death (Parsi, 2001; Sack et al., 2004, CDC, 2013). The symptoms of cholera can last for 4 to 5 days, but this time can be shortened to 2 or 3 days with antibiotics. In endemic areas, children under 5 years of age make up, it is estimated, approximately half of all cases, but all age groups are affected by cholera and can have high mortality, whether in epidemic or endemic areas (Ali, et al., 2012; Sack, et al., 2004; WHO, 2012). Without treatment, there is a 50 to 60% mortality rate, but if patients receive both fluid and electrolyte replacement, the mortality rate can be under 1%; any mortality rate greater than 1%, because the disease can be easily treated, is considered unacceptable and indicative of inadequate health care (Murthy, 2013; Murray, et al., 2009).
Excretion, Hospitalization Rates, Immunity
If demonstrating symptoms, an individual can shed before and for up to two weeks after the resolution of symptoms; the number of organisms present in feces immediately after passage can be between 10^7 and 10^9 organisms/mL, and these organisms can be in a hyper-infectious state, lowering the dosage needed for illness by 10 to 100 times. The infectious dose of V. cholorae is approximated to be anywhere between 10^3 and 10^8 organisms, depending on such factors as contaminated water exposure, contact with infected individuals, and personal health status (Harris, LaRocque, Qadri, Ryan & Calderwood, 2012; Morris, 2011; Public Health Agency of Canada, 2011).
Between October 2010 and October 2012, the cholera hospitalization rate in Haiti was approximately 54.5% (329,697 hospitalizations out of 604,634 reported cases) (Barzilay, et al., 2013). The hospitalization rate in the Indian jute mill was 51% (approximately 101 of 197 identified cases) (Mridha, et al., 2011).
Infection with the organism can provide some long-term protection against future infection (Harris, 2010). Although there are 2 major vaccines available, no vaccine has been shown to provide significant long-term protection (CDC, 2013; Murray, et al., 2009).
Exposure and Environments
This organism is naturally found where rivers and seas meet (estuaries) and in marine environments; it has also been found in freshwater environments in endemic areas (Parsi, 2001; Mishra, Taneja, & Sharma, 2011; Murray, et al., 2009). Humans can become infected if they come in contact with these areas, or they eat or drink food or water contaminated by feces containing V. cholera, as cholera is generally transmitted by fecal-oral transmission. In endemic areas and in epidemics, contaminated water is the main source of the organism. Food may be contaminated by contaminated water, then the food could provide a source of the infection (Parsi, 2001; Murray, et al., 2009; Sack, Sack, Nair, Siddique, 2004). The number of these organisms in the environment can increase due to warmer temperatures, higher pHs (6.5 – 9.0), higher salinity, and the presence of algal blooms; V. cholera does show seasonal variation in growth (Mishra, et al., 2011; Huq, Sack, Nizam, Longini, Nair, Ali, et al., 2005). V. cholera is found in association with certain species of plankton that can protect it from being treated by chlorine (de Magny, et al., 2011; Murray, 2009).
The concentration of naturally-occurring cholera in surface waters is generally low (Parsi, 2001; Wang, Xu, Deng, Chen, Li, et al., 2010). One study of areas in Peru and/or Bangladesh, suggested that, in these endemic areas, the number of infectious V. cholera organisms naturally occurring in “pristine waters” is 10 to 100 CFU/mL, while at the peak of an epidemic in these same areas, in heavily sewage-contaminated waters, the concentrations can be approximately 100,000 CFU/mL (Morris, 2011). Another study of an endemic region in India demonstrates the amounts and strains of cholera in freshwater, for example, depend on whether the area is endemic for cholera, if there is an outbreak occurring at the time of sampling, the environmental conditions present, and whether the water being tested is freshwater, brackish, or seawater; a second study in Bangladesh confirms similar results for coastal waters (Mishra, et al., 2011; Huq, et al., 2005). This variation in endemic areas could reflect a greater difficulty in detection in non-endemic waters, at least initially, during an epidemic.
A comparison of the outbreak in Haiti to the infected jute mill workers in West Bengal India is difficult because of the different spatial and temporal scales and different exposure mechanisms. In the jute mill, we can assume all workers drank the infected water because the outbreak was confined to the jute mill itself. We cannot assume that all people in Haiti have been exposed to the contaminated water of the Artibonite. Unlike in the jute mill, the means of transmission in Haiti is changing over time (Gaudart et all. 2013). To facilitate a comparison to the jute mill, the exposure assessment is limited to the lower Artibonite region of Haiti. The initial outbreak in Haiti, along the Artibonite River, occurred in six communes: St Mar, Dessalines, Petite Riviere, Grande, Saline, Verrettes, and Desdunes downstream from the camp at Mirebalais, where the first few cases were reported. There was initially some disagreement whether Cholera was transmitted by road or the Artibonite River because the concentration of Cholera may not have been high enough to cause infection, but statistical analysis strongly suggests the Artibonite was the source of Cholera outbreak (Piarroux et al. 2011).
Even after limiting the analysis to the Lower Artibonite, uncertainty is high. Some people in the communes may have had private water access. (Doucet 2013). This might cause bias as well as increase uncertainty if a large segment of the population was not exposed to the Artibonite, yet were considered in the dose response. Using too high a population in the dose response model would lower the calculated dose. Because the exact number of people that were or were not exposed to water from the Artibonite is unknown, as is the quality of private water, we cannot conclusively state results are biased, but believe they are.
Another uncertainty factor is our lack of primary sources. Piarroux et al. reported that by October 22nd there were 3020 cholera cases and 129 deaths in the six communes we are modeling. The Pan American Health Organization (PAHO) situation report #5 stated there were 3769 cases and 284 deaths, 96% of which were in Artibonite (PAHO). Because reports from Artibonite after October 22 may include cases caused by secondary transmission, we are using the 3020 cases reported in Piarriox et al. Unfortunately we have no way to determine what percentages of people were infected from the Artibonite River and from secondary transmission. It is likely that not all cases were reported by October 22nd, and a second source of bias may be underreporting, which would also decrease the calculated dose. Lack of data is a major source of uncertainty, and our decision to only use cases reported before the 22nd may bias the results.
Several dose response models were used to compare the outbreak in the jute mill in Kolkata, West Bengal, India with the Lower Artibonite region in Haiti during the first few days of the 2010 outbreak. Because we do not have empirical data of the concentrations of cholera, we are using population information for the six affected municipalities: St Mar, Des Salines, Petite Riviere, Grande, Saline, Verrettes, and Des Dunes, and all males of working age, 15-64, in India (the only population affected in the jute mill outbreak) (Mridha, et al., 2011).
By assuming all Haitians in these municipalities and all Indian males 15-64 consumed infected water, using four different beta-Poisson dose response models we calculated the expected number of organisms consumed per person. Population for the municipalities was taken from the French Wikipedia pages for Haiti because the English pages were missing data. Population of the Indian residential colony around the jute mill was 5910 individuals, but only 197 male individuals between the ages of 15 and 64 were affected, a subset of a population of approximately 4,419 people (Mridha, et al., 2011).
The first dose response model is taken from the available models in the wiki. It is the recommended model, but is calibrated for the Inaba strain, not the Ogawa strain responsible for the outbreak in Haiti. Additionally, it is calibrated for infection and a pH buffer. Figures for both India and Haiti are for illness, not infection. We assumed 25% of people infected would show symptoms (WHO,2012). The water consumed in Haiti and India was likely not buffered, and we have no evidence that the local diet contained food that would neutralize stomach acid. Because of these assumptions, we believe this is not the best model. Our calculations shown in Table 2 support this, showing the average number of organisms consumed per person in Haiti would be 1.1.
The second and third models are from the previous workshop's dose response for Cholera in Haiti. The second model is a modified version of the recommended model, calibrated for illness rather than infection. The third model is for the El Tor biotype. Unfortunately the third model’s reference is missing, so its’ accuracy cannot be verified. Both models also calculated very low numbers of organisms consumed.
The fourth model is for the Inaba strain, but is calibrated for illness and does not use a pH buffer. It is likely the most accurate. Results from table 2 confirm this, with results for the number of organisms consumed in the typical range of 10+e4 to 10+e8.
|Reference||Host||Agent Strain||Route||nDoses||Dose Units||Response||Best Fit Model||&alpha||N50|
|Haas, C N., Rose, J. B. And Gerba, C. P (1999)||human||Inaba 569B||ingestion with PH buffer||6||CFU||infection||beta-Poisson||2.50E-01||2.43E+02|
|Haas, C N., Rose, J. B. And Gerba, C. P (1999)||human||Inaba 569B||ingestion with PH buffer||6||CFU||illness||beta-Poisson||4.90E-01||N50= 3.365E+03 }}|
|Stott et al. (previous workshop)||Human||Inaba N16961 (El Tor)||Ingestion with PH buffer||7||CFU||illness||beta Poisson||1.690E-01||3.365E+03|
|Hornick et al. 1971 (WIKI recommended models)||Human||Inaba N569B (El Tor)||Ingestion without PH buffer||7||CFU||illness||beta Poisson||1.980E-01||6.36E+08|
The extreme discrepancy in the number of ingested organisms consumed shown in table 2 highlights the importance of choosing the correct model. The fourth model is likely the only valid model. Unfortunately, we do not have a dose response model for illness, without a ph buffer, for the El Tor biotype.
Results in table 2 also indicate using a pH buffer is the most important factor in the calculation. It would be interesting to conduct a sensitivity analysis with more models. Because the difference in calculated number of organisms consumed between pH and non pH models is several orders of magnitude greater than differences related to other criteria, we believe our recommendation is valid without a sensitivity analysis.
Haas, C N., Rose, J. B. And Gerba, C. P (1999). Quantiative microbial risk assessment. John Wiley and Sons, ISBN 0-471-18397-0 from http://qmrawiki.msu.edu/index.php?title=Case_Study%3A_Cholera#tab=Dose_Response
Hornick, R.B., Music, S.I., Wenzel, R., Cash, R.A., Libonati, J.P., Snyder, M.J., Woodward, T.E. (1971) The Broad Street Pump Revisited: Response of Volunteers to Ingested Cholera Vibrios Bulletin New York Academy of Medicine 47(10): 1181-1191 Free Full Text via NIH from http://qmrawiki.msu.edu/index.php?title=Vibrio_cholerae%3A_Dose_Response_Models#1
Mridha, P., Biswas, A.K., Ramakrishnan, R., & Murhekar, M.V. (2011). The 2010 outbreak of cholera among workers of a jute mill in Kolkata, West Bengal, India. Journal of Health, Population and Nutrition 29(1), 9-13. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3075051/ Stott et al. from http://qmrawiki.msu.edu/index.php?title=Case_Study%3A_Cholera#tab=Dose_Response
"Départements d'Haïti - Wikipédia." (2014) from http://fr.wikipedia.org/wiki/D%C3%A9partements_d%27Ha%C3%AFti#D.C3.A9partement_de_l.27Artibonite.
World Health Organization (WHO). (2012, July). Media centre: Cholera: Fact sheet no 107. Retrieved from http://www.who.int/mediacentre/factsheets/fs107/en/
Risk characterization is complicated by the difficulty in choosing an appropriate dose response model and the high uncertainty in the Haiti models. Dose response model 3 is the only one designed for the El Tor biotype, but it appears to be the least accurate model, with the lowest reported number of organisms. Only model 4, which was not calibrated with a pH buffer, yielded realistic results.
Risk characterization should be different for Haiti and India. Because the outbreak was contained within the jute mill in India, there was no risk to the outside population (Mridha et al. 2011). In Haiti, we should be concerned with changing modes of transmission and how the infection will continue to spread (Guadart et al. 2013). For comparison purposes, we have limited the analysis to the Lower Artibonite region, in the days just following the outbreak. This study does not attempt to quantify the future risk for Haitians.
The attack rate for Haiti, in the area and time the dose response was limited to, is 0.4 illnesses per 100 people. See the exposure assessment section for details on how this figure was calculated. For working aged males in the Indian jute mill the attack rate is 8.9. Obviously comparing the attack rates from the initial Haitian outbreak with the Indian outbreak is grossly misleading. Over 600,000 Haitians have been infected and the outbreak has become endemic. Table 1 below shows the attack rates for Haiti from 2010-2012.
Using the most likely dose response model from this study, calculations show the average dose ingested by Haitians in the Lower Artibonite region was 29 times higher than that ingested by Indian workers. During the initial stages of the epidemic in the Lower Artibonite region, the case fatality rate (CFR) was 4.3 deaths per illness, compared to 0 in the Indian jute mill. Dose does not appear to be a good predictor of the likelihood of death or that a disease will spread. The data is table 1 confirms this. Using SPSS 20, normality tests were run on the attack rates and CFR’s. Because CFR’s strongly rejected all normality tests, non-parametric correlation coefficients were calculated instead of the Pearson’s correlation coefficient. The Spearman’s correlation coefficient is -.664 and the Kendall’s is -.545. Both values are statistically significant at the .05 level, indicating attack rates and CFR's are negatively correlated.
Factors other than the initial dose appear to be the primary risk. Obviously the most important factor is containing the Cholera infection. This was not possible in Haiti, but was accomplished in the jute mill. Further research into the inverse correlation between attack rates and CFR’s would be interesting. Perhaps it would provide information to help direct treatment.
|Port Au Prince||13.32||0.64|
Gaudart J, Rebaudet S, Barrais R et al. Spatio-temporal dynamics of cholera during the first year of the epidemic in Haiti. PLoS Negl Trop Dis 2013; 7(4): e2145
Mridha P, Biswas AK, Ramakrishnan R et al. The 2010 outbreak of cholera among workers of a jute mill in Kolkatta, West Bengal, India. J Health Popul Nutr 2011; 29:9-13
Pan American Health Organization (23 December 2013). "Epidemiological Update, Cholera". Retrieved 31 January 2014 from http://www.paho.org/hq/index.php?option=com_content&view=article&id=1239&Itemid=2291 and Wikipedia
According to World Health Organization estimates “officially reported cases [of cholera] represent only 5-10% of the actual number occurring annually worldwide” (Ali, Lopez, You, Kim, & Sah, 2012). This statement reflects that, even in areas where cholera is familiar to health care professionals, its full health impact and burden are unknown. The inadequate information regarding numbers of afflicted can be due to a multitude of factors that may include inadequate surveillance and reporting, lack of recognition of milder cases, and the inability of public health officials to reach populations effectively or adequately in trying to evaluate and prevent cholera.
The diagnosis of cholera presents a challenge as quick methods are still in development. Using microscopy is not recommended because cholera organisms look like other enteric organisms, so culture on special media provides a better diagnostic tool (Murray, et al., 2009; CDC, 2013a; Lamond & Kinyanjui, 2012). The use of lab diagnosis is still required for confirmation of cholera cases, and this can present a challenge to areas with inadequate facilities or facilities that are not conveniently available. Fortunately, in an outbreak, due to the potential volume of patients not every patient must be confirmed by culture (CDC, 2013a; Lamond & Kinyanjui, 2012). In the case of an outbreak, once confirmed, using the following clinical diagnosis (in this case, in the Haiti cholera epidemic) is sufficient to establish infection, according to the World Health Organization (WHO): “In an area where there is a cholera epidemic, a patient aged 5 years or more develops acute watery diarrhea, with or without vomiting.” (WHO, 2014b).
One of the most important initial public health responses for individual patients lies in the treatment of cholera. “The first 24 hours of cholera manifestation are the riskiest, and if the sufferer is not rehydrated, death can result”; with appropriate and timely treatment, case fatality rates can be kept below 1%. A case-fatality rate higher than 1% reflects inadequate health care in an area (Lamond & Kinyanjui, 2012; Murthy, 2013). Treatment of individuals is mainly provided with fluid and electrolyte replacement; there is no effective long-term vaccine for the prevention of cholera. There are vaccines available, but there are concerns with their effectiveness, adverse reactions, and the need for two separate doses; vaccines are only considered a supplementary public health tool to the main management techniques to be discussed below (WHO, 2014a). An important limitation to effective treatment in non-endemic areas is the ability (or inability) of doctors and nurses to recognize cholera cases when they have limited to no experience with its presentation or symptoms. Comparing Haiti’s outbreak to the outbreak in the jute mill in Kolkata, India around the same time, the ability to recognize cholera and quickly control its spread lead to a quick resolution of the cholera outbreak in Kolkata, while a poor infrastructure coupled with the unfamiliarity with a non-endemic disease seemingly has prolonged the presence of cholera over years in Haiti (Mridha, Biswas, Ramakrishnan, & Murhekar, 2011).
Risk Factors & Personnel
Major risk factors for the transmission and spread of cholera include “[p]oor social and economic environment and unstable living conditions,…[u]nderlying diseases and conditions…, [g]ender…, [and] [e]nvironmental and seasonal factors” (Lamond & Kinyanjui, 2012). Specific circumstances of concern pertinent to Haiti included refugee camps, insufficient water supply, and poor sanitation—all of these presented due to a severely damaged infrastructure and massive displacements of individuals, which, as seen previously, complicated exposure assessment, dose-response modeling, and risk characterization due to secondary transmission. In managing cholera, not only is it important to have adequate resources to treat individuals promptly and effectively, there must be great consideration as to what physical infrastructure is available to provide what communities need. Trained health officials and professionals who can recognize cholera's presence in patients and in communities are required so as to provide the appropriate care, and clean water sources must be sufficient in number to prevent further infection.
Community prevention methods should be encouraged through local community leaders, of both genders, who are trustworthy to the communities involved. Although two main sources of cholera were posited: UN aid workers coming from Asia or naturally occurring cholera, the growth of which was encouraged by the El Nino phenomenon, other local ideas about such individuals as voodoo practitioners being the cause were also suggested. Such a variety of explanations creates for uncertainty and can cause instability due to violent responses (Lamond & Kinyanjui, 2012; Valme, 2010; Frerichs, Keim, Barrais, & Piarroux, 2012). It is important, then, that community members not only have a trustworthy source of information that is familiar to them but also that public health officials have individuals working to help relieve tensions between groups due to rumors, help guide individuals to the appropriate interventions, and to relay concerns and problems presented by communities affected by the epidemic (Lamond & Kinyanjui, 2012; Valme, 2010; Frerichs, et al., 2012). These local contacts can further aid public health officials in their evaluation and prevention efforts because the communication between communities and public health officials in a cholera epidemic involves both specific interventions and constant evaluation of affected areas.
At the beginning of an outbreak, there should be a rapid initial evaluation of where cholera cases are occurring and spread, available treatment facilities and staff, community demographics, dose-response, attack rates, case-fatality rates, location and quality of water sources, and location of latrines. During and after this initial report's construction, efforts should be focused on education regarding prevention and intervention strategies. Intervention methods include, if no water treatment facility is available, treating water with chlorine tablets or household bleach (2 drops per liter or 8 drops per gallon) and keeping water covered in a clean container, frequent hand-washing with soap, burying feces (preventing defecation in water sources), cooking food well, and cleaning selves and clothes 30 meters away from drinking water sources. It is important to emphasize the need for rehydration and treatment for the severely ill. Concerns from community members regarding chlorine usage, impacts on their water usage, resource availability, and suspicions and fears they have about health care, their homes, or water treatment must also be elicited, so as to allow for compromises that can achieve equally effective prevention and health care (e.g. instead of soap and water, ash/sand and water can be used as the rubbing action is effective in helping to improve hygiene). Evaluations and reports need to occur frequently to ensure that methods of prevention and treatment are working effectively; in the case of Haiti, it is particularly important to ensure that individuals have safe, clean sources of water, as the evaluation of the outbreak revealed that the spread of cholera followed the course of a river and along the coastlines as individuals used the water and as they attempted to move out of cholera affected areas. These prevention methods need to be spread quickly and constantly presented to all affected areas, making community outreach workers even more important (Lamond & Kinyanjui, 2012; CDC, 2013b).
While the patient and water treatment methods are relatively cheap, the major limitations to the prevention and treatment methods are availability of resources, quality of outreach with respect to the population, poor infrastructure, and lack of enough money to afford the resources for large groups of people. While it is necessary to have appropriately trained health care staff, it is also important to have trained individuals and easily legible materials at or near water sources informing people as to appropriate prevention methods (Lamond & Kinyanjui, 2012). This information must be appropriate to the literacy levels of the population; in Haiti, the literacy rate, as of 2006, was approximately 48.7% of the population over the age of 15 years, suggesting pictoral explanations or verbal communications (e.g. radio) may be more appropriate for such a population (CIA, 2013). Other considerations for a population is the number of individuals who have access to what are termed improved sanitation facilities (closed facilities with or without piped sewer systems that connect to a larger system)—Haiti, as of 2010, only had 17% of the population with such a living situation, requiring education regarding where to place latrines or how to properly dispose of human waste (CIA, 2013). Finally, religious considerations and political stability also compromise or aid relief, prevention, and control efforts, as previously mentioned in regard to voodoo and perceptions of international authorities.
Global Health Implications of Cholera in Haiti
Out of Haiti’s cholera epidemic came two very important global public health implications: (1) Haiti, a region non-endemic for cholera, now is endemic for the disease, and (2) international relief and care efforts can cause harm in an effort to do good. Haiti had 100 years without cholera prior to the epidemic that occurred secondary to an earthquake; now, the country and the international community must adjust to managing an old disease in a new country that had no preparation for it. The international community’s response, though, may face great barriers to improvement, as it appears that UN workers were the ones possibly responsible for bringing in the disease from Asia and contaminating a Haitian water source as a result of the UN camp’s poor sanitation system (Frerichs, et al., 2012). Communication and trust must now be rebuilt as the country of Haiti also physically rebuilds itself, slow processes that require careful attention by all parties involved. The risk management strategies mentioned here allow for such rebuilding to occur, but their appropriate execution and success depends upon the care, consideration, and equitable communication between all affected and working to improve such a damaged nation.
- Nepalese origin of cholera epidemic in Haiti, , Clinical Microbiology and Infection, Volume 18, (2012)
- Spatio-temporal dynamics of cholera during the first year of the epidemic in Haiti, , PLoS neglected tropical diseases, Volume 7, p.e2145, (2013)
- Understanding the cholera epidemic, Haiti, , Emerging infectious diseases, Volume 17, p.1161, (2011)
- Cholera surveillance during the Haiti epidemic—the first 2 years, , New England Journal of Medicine, Volume 368, p.599–609, (2013)
- The 2010 outbreak of cholera among workers of a jute mill in Kolkata, West Bengal, India, , Journal of health, population, and nutrition, 2011, Volume 29, p.9, (2011)
- Medical microbiology: Mosby Inc, , (2009)
- The incubation period of cholera: a systematic review, , Journal of Infection, Volume 66, p.432–438, (2013)
- The global burden of cholera, , Bulletin of the World Health Organization, Volume 90, p.209–218, (2012)
- Cholera—modern pandemic disease of ancient lineage, , Emerging infectious diseases, Volume 17, p.2099, (2011)
- Environmental and epidemiological surveillance of Vibrio cholerae in a cholera-endemic region in India with freshwater environs, , Journal of applied microbiology, Volume 112, p.225–237, (2012)
- Quantitative Microbial Risk Assessment, , New York, NY, (1999)
- The Broad Street pump revisited: response of volunteers to ingested cholera vibrios, , Bulletin of the New York Academy of Medicine, Volume 47, p.10, (1971)
- Cholera outbreak guidelines: Preparedness, prevention and control, , (2012)